Assessment of the Economic Impact of the Saiga Population on Pasture Ecosystems and Agriculture

Abstract

This article discusses the issue of pasture degradation in the West Kazakhstan region, which has been caused by a significant rise in the population of saigas (Saiga tatarica). This study aims to quantify agricultural losses and establish the relationship between saiga numbers and ecosystem changes. The research methodology incorporates field observations, agroecological observations, and mathematical analysis. Correlation and regression analyses were performed. The application of correlation and regression models confirmed a statistically significant relationship between the growth of the saiga population and the decline in hay yield: increasing animal numbers lead to a reduction in pasture productivity. The greatest losses were recorded in the Bokeiordinsky district, where grazing pressure exceeded 62.5%. It was concluded that urgent measures are needed to regulate the saiga population, restore degraded land, and introduce rotational grazing. Mechanisms for the adaptive management of wild populations that take into account the interests of the agricultural sector were proposed. The findings obtained can provide a solid foundation for informed decision-making in agricultural and environmental policy within the region.

1. Introduction

The conservation of ecosystems and biodiversity in the 21st century is acquiring particular significance amid global climate change and increasing anthropogenic pressure. In the steppe and semi-desert regions of Central Asia, the problem of balancing agricultural development with nature conservation is becoming increasingly acute.

Maintaining a balance between agricultural development and biodiversity conservation is challenging for sustainable land use in the steppe and semi-desert regions of Central Asia. The rapid increase in saiga (Saiga tatarica) numbers in Western Kazakhstan has elicited conflicting reactions from farmers, environmental organizations, and government agencies. On the one hand, saiga are protected internationally under the Convention on Migratory Species [1]. On the other hand, high local densities can reduce the productivity of fodder resources and accelerate pasture degradation, with implications for food security.

The conflict between wild ungulates and agriculture is a global issue. In Eastern Africa, research demonstrates substantial economic losses caused by elephants (Loxodonta africana) and African buffalo (Syncerus caffer), which compete with livestock for forage resources [2]. Similarly, studies in Mongolia have examined the impact of wild ungulates, including the Mongolian saiga (Saiga tatarica mongolica), on pasture ecosystems and livestock production [3].

Modern approaches to managing wild ungulate populations are grounded in adaptive management, spatial monitoring, and economic impact assessment. For the saiga, a keystone species of Kazakhstan’s steppe ecosystems, it is important to account for both ecological functions and the risks of pasture degradation.

In Western Kazakhstan, the rapid growth of saiga populations has raised concerns among farmers and conservationists. The central question of this study is: how does saiga abundance affect the forage productivity of pasturelands and livestock productivity?

Although there is a substantial body of research on human–wildlife conflict in Africa and Mongolia, Kazakhstan is represented only fragmentarily in the international literature. Existing studies rarely combine saiga demographic data with quantitative assessments of agroecological losses, which limits the effectiveness of policy in the region.

In scientific literature, the issue of interaction between wild ungulates and agricultural systems is studied in the context of ecosystem conflicts [4], economic assessment of environmental losses [5] and spatial analysis of the load on forage lands [6]. However, most studies focus on Africa, Mongolia and Eastern Europe, while Kazakhstan is represented only fragmentarily. This creates a gap in the international scientific field and limits the effectiveness of agricultural policy, which must take into account both environmental and economic factors.

Numerous studies document a rise in human–wildlife conflicts [2,7]. Modern wildlife population management emphasizes integrative strategies that combine adaptive management, spatial modeling, and economic impact assessment. Adaptive management entails iteratively adjusting decisions based on monitoring population status and evaluating management effects [8]. Methods that quantify ecosystem services and agricultural damages caused by wildlife are widely used [9]. Effective policies must weigh both benefits, such as environmental and recreational values, and costs in human-dominated landscapes.

Population management, as defined by Hone et al. [10], involves intentional human intervention in the dynamics of wild animal populations to achieve a stable population while preserving their ecological functions. However, traditional methods that focus on the exclusion of specific age or sex groups often fall short in managing rapidly growing populations with high reproductive rates. Servante et al. [11] emphasize that these methods do not provide a sustainable solution and highlight the need for integrated management that considers the demographic structure and reproductive dynamics of populations.

Methods for the economic assessment of damage caused by wild animals are actively evolving in international practice. In North America and Europe, compensation models for farmers affected by large herbivores are widely implemented [12].

These issues are particularly important concerning Saiga tatarica, which is a crucial component of the steppe and semi-desert ecosystems in Kazakhstan. According to UNEP/WCMC [13], saigas are a key species for maintaining biodiversity. They help shape the structure of plant communities and provide food for predators and scavengers, thereby contributing to the sustainability of these ecosystems.

To effectively regulate population size, it is important to evaluate its dynamics within the context of the species’ biological capabilities. Hone et al. [10] highlight that the maximum annual population growth rate (rm) is a crucial parameter for developing management strategies. This metric helps predict how populations can recover and establishes acceptable levels of animal removal without jeopardizing the sustainability of the population.

The spatial structure of populations plays a crucial role in their management. Hampton et al. [14] point out that the movement of individuals between local groups can undermine local population control efforts if we do not consider genetic connectivity and migration patterns between populations.

In environments characterized by high variability in external factors, effective management should adopt adaptive approaches. This is supported by the concept of dynamic capabilities proposed by Teece, Pisano, and Shuen [15], which highlights the need for a flexible response to changes in the external environment. Manlik et al. [16] emphasize that population management strategies must consider demographic variability and environmental uncertainty, focusing on realistic management scenarios rather than solely relying on theoretical models.

Effective management of saiga populations requires integrating demographic analysis, spatial monitoring, and adaptive population management methods. This approach aims to achieve a sustainable balance between conservation of the species and minimizing damage to the agricultural sector [4].

Kazakhstan is crucial for the global conservation of saiga, as most remaining populations reside within its borders [17]. The international community actively supports Kazakhstan’s initiatives to protect and restore saiga populations, highlighting the importance of international cooperation in these efforts [18].

Saiga antelopes (Saiga tatarica) play a vital ecological role in the steppe and semi-desert ecosystems of Kazakhstan. Currently, two species of saiga are recognized by international conventions: S. tatarica, which is found in Kazakhstan, Uzbekistan, and Russia, and S. borealis, which inhabits Mongolia. However, several researchers, including those from the IUCN, consider them to be subspecies of S. tatarica. They differentiate between S. t. tatarica, which is widespread in Russia, Kazakhstan, and Uzbekistan, and S. t. mongolica, native to Mongolia. It is believed that both subspecies once inhabited China [19].

Four populations of S. tatarica have been identified. Three of these populations primarily reside in Kazakhstan, while the fourth is located in the northwestern Caspian region of Russia. Currently, Kazakh populations account for over 90% of the total number of this species. Among these are the Betpakdalin population, which inhabits central Kazakhstan and sometimes crosses into Russia; the Ural population, found in northwestern Kazakhstan and sharing a cross-border area with Russia; and the Ustyurt population, located in southwestern Kazakhstan, which occasionally crosses the border into Uzbekistan [13].

Saigas act as “ecosystem engineers,” significantly influencing plant communities and creating favorable conditions for other animal species, including rodents, insects, and birds [20]. Their grazing helps maintain plant diversity and prevents the dominance of any single plant species, contributing to the overall balance of the ecosystem.

A literature review shows that previous studies have primarily considered the environmental impact of wild animals in isolation from economic losses. Meanwhile, quantitative models integrating demographic, spatial and agronomic factors have hardly been used in the case of saigas. Furthermore, the studies did not emphasize the direct link between saiga population density and the decline in fodder crop productivity within specific administrative districts. This article proposes a methodology that fills these gaps by combining statistical models with real field observation data.

The present study provides an initial assessment of the direct statistical relationship between the size of the saiga population and pasture productivity, deliberately excluding climatic variables and livestock population dynamics. This represents a methodological limitation of the first analytical stage, aimed at identifying the pure structural effect of grazing pressure exerted by saiga. Climatic factors and anthropogenic load constitute the focus of subsequent phases of the research; however, the present analysis applies a deliberately simplified model.

Such a design is consistent with methodological approaches employed in the works of Gren et al., Manlik et al. [5,16] and other authors referenced in the literature, where preliminary impact assessment is based on a minimal set of explanatory variables. The choice of a simplified specification ensures full transparency of the model structure and enables an unambiguous interpretation of the influence of the key factor—saiga population size—without confounding effects of climate or livestock management. This approach ensures the validity of the first-stage findings and establishes a solid foundation for further model expansion. The results allow for an assessment of the ecological and economic consequences of saiga population growth and inform recommendations for sustainable management policy.

2. Research Methods

The study uses a quantitative approach based on the case study method. Three representative areas of the West Kazakhstan region were selected, where the saiga population has the greatest impact on the agricultural sector. The methodology combines field empirical observations, agroecological measurements, correlation and regression analysis, as well as an economic assessment of losses in crop yields and livestock productivity.

Field studies were conducted in 2024 to assess the state of pastures in the Bokeyordinsky, Zhanibeksky, and Kaztalov districts of the West Kazakhstan region. The main biometric indicators of pasture ecosystems that were examined include projective vegetation cover, herbage height, green mass yield, herbage composition, and the energy value of pasture feed.

The statistical analysis involved correlation analysis to evaluate the relationship between the number of saiga antelopes, pasture and livestock productivity. This was achieved using a correlation matrix. Regression analysis was then applied to construct models quantifying the effect of saiga numbers on hayfield and pasture productivity. Separate regressions were estimated for each district. Model adequacy was assessed using the coefficient of determination R2R2 and the Fisher test. The economic component evaluated damage by calculating productivity losses on forage lands and by analyzing changes in sheep live weight across the study areas.

A quantitative design was selected to numerically evaluate links between saiga population size and crop and livestock productivity. The case-study approach was used because impacts vary geographically: districts in Western Kazakhstan differ in both pasture degradation and saiga abundance. Field observations documented the current condition of pastures and standing vegetation. Agroecological measurements (e.g., grass height, projective cover, productivity, and feed energy content) enabled comparisons across plots experiencing different grazing pressures].

Correlation analysis was employed to identify statistical dependencies between the saiga population and agricultural productivity indicators. Regression analysis was then used to create mathematical models showing the impact of the saiga population on crop productivity in specific regions.

The STATISTICA 12 software package was employed for data processing and analysis. The assessment of the load on pasture lands was conducted according to the maximum permissible standards set forth by the Order of the Ministry of Agriculture of the Republic of Kazakhstan, dated 14 April 2015, No. 3-3/332 [21]. The analysis of the actual availability of pasture resources involved calculating the shortfall in areas needed for livestock and saigas.

The agroecological assessment of the impact of saiga antelopes on the sustainability of pasture ecosystems was carried out based on the analysis of key indicators of vegetation condition. To address this objective, the following measurements and field observations were conducted at designated monitoring sites: changes in the species composition of the grass stand; assessment of forage biomass yield; evaluation of pasture forage quality.

A linear regression model was used to analyze the effect of saiga abundance on hay productivity due to its simplicity, interpretability, and adequate accuracy given the available data. This choice followed a comparison with polynomial alternatives, where the linear specification showed more stable performance on test subsamples and clearer coefficient interpretation. Prior to estimation, the assumptions underpinning the Gauss–Markov theorem were checked to ensure the validity of ordinary least squares. Multicollinearity was evaluated using variance inflation factors (VIF), which remained below the conventional threshold (VIF < 5), indicating acceptable independence among predictors. Residual heteroscedasticity was tested with the Breusch–Pagan procedure and was not detected at conventional significance levels (p > 0.05). Together, these diagnostics support the assumption of constant residual variance and the suitability of the linear model for inference.

The study used the following variables:

Dependent variable (Y)—hay productivity, measured in kilograms per hectare (kg/ha);

Independent variable (X)—number of saigas (per 1000 heads) in the respective administrative district.

Data sources included: official statistical compendia of the Republic of Kazakhstan (“Agriculture, Forestry and Fisheries in the Republic of Kazakhstan, 2017–2023”; “Environmental Protection and Sustainable Development of Kazakhstan, 1981–2023”) [22,23], materials from field observations and measurements conducted under the IRN AP23486846 research project (2024) [24], and UNEP/CMS analytical documents.

Key variables included:

—

The abundance of saiga in the Bokeyorda, Zhanibek, and Kaztal districts;

—

The area of available and required pastures (in hectares, according to the standards set by the Order of the Ministry of Agriculture of the Republic of Kazakhstan dated 14 April 2015, No. 3-3/332) [21];

—

Hay productivity (kg/ha), derived from biometric measurements and farm statistics.

The same linear regression model was constructed for all three districts. The model was estimated both in its original (unstandardized) form and in standardized form. The standardized specification allows for a correct comparison of the effect size across different territories.

Unstandardized model (in original units):

$$Y_t={\beta }_0+{\beta }_1\cdot Saig{a}_t,$$

(1)

where:

$Y_t$—pasture productivity (kg/ha);

$Saig{a}_t$—population size of the Ural saiga population (number of individuals);

$t$—year of observation.

Standardized model (model with β_s). After standardizing both variables, the equation takes the form:

$$Z_{Y,t}={\beta }_s\hspace{0.17em}{Z}_{Saiga,t},$$

(2)

where:

$Z_Y$—standardized pasture productivity;

$Z_{Saiga}$—standardized saiga population size;

${\beta }_s$—standardized regression coefficient (Standardized Beta).

The constant term is omitted because the means of the standardized variables are equal to zero.

Comparative statistical results (calculations performed in SPSS 19). Summary table of standardized coefficients β_s and their statistical significance

Table 1. Standardized regression coefficients (β_s) and their statistical significance across districts. (2010–2024).

District	βs	t	p-Value	Interpretation
Bokey Orda	−0.927	−6.967	0.000	Very strong negative effect
Zhanibek	−0.861	−4.784	0.001	Strong negative effect
Kaztalov	−0.888	−5.459	0.001	Very strong negative effect

.

The standardized coefficient β_s indicates how strongly pasture productivity (measured in standard deviations) responds to a one-standard-deviation change in the size of the saiga population. Values of |β_s| greater than 0.8 are classified as very strong effects.

Across all three districts, the relationship is negative, statistically significant (p ≤ 0.001), and very strong in terms of effect size. This indicates that the increase in saiga population exerts a pronounced, persistent, and systematic impact on pasture degradation. These findings are crucial for informing wildlife management policy and for developing measures aimed at mitigating pasture degradation.

The model employs a single population size indicator for the Ural saiga population across all three districts. This approach is justified by the fact that the Ural population constitutes a unified migratory herd, with groups of animals moving continuously across the Bokeyorda, Zhanybek, and Kaztal districts throughout the year. Administrative boundaries do not correspond to natural migration routes, and grazing pressure on pastures is generated by the total biomass of the population rather than by isolated local groups.

Therefore, a single population indicator accurately reflects the regional level of grazing pressure exerted simultaneously across all three territories.

Field data on the saiga population size and pasture condition were collected as part of the IRN AP23486846 research project ‘Economic assessment of the impact of wild animal populations (saigas) on the agricultural sector and ways to reduce the damage caused’ [24]. The project monitored saiga numbers in the Bokeiordinsky, Zhanibeksky and Kaztalovsky districts using ground-based visual counts and georeferencing of habitats. In addition, biometric indicators of pasture quality (e.g., productivity, grass height, cover and feed value) were recorded in areas with controlled and free grazing. These data formed the basis for constructing regression models in the article. Analytical reports from the United Nations Environment Programme/Convention on Migratory Species [13] and the latest edition of the Memorandum of Understanding on the Conservation of Saigas [18] were used to verify the dynamics of the saiga population and its range distribution. These documents officially confirmed the importance of the saiga population in Kazakhstan, which accounts for over 90% of the global population of the species.

The data collected enabled a comprehensive evaluation of the degradation processes affecting pasture ecosystems due to the increasing population of saiga antelopes. Additionally, this analysis helped in developing recommendations to minimize economic losses in the agricultural sector.

3. Results

Saigas play a crucial role in the food chain by providing nutrition to large predators such as wolves and scavengers. This relationship supports the health of these predatory populations and prevents the unchecked growth of saiga numbers [25].

Saigas are a crucial component of steppe ecosystems and play a significant role in maintaining biodiversity. However, their interaction with agriculture, particularly in areas with intensive land use, can result in substantial economic losses. In the West Kazakhstan region, where agriculture is a key sector of the economy, the impact of saigas on pastures and hayfields is becoming an increasingly pressing issue.

The presented map highlights the territory of the West Kazakhstan region (WKO), where red lines mark the administrative districts that concentrate the main part of the saiga population. Figure 1. The map illustrates the key zones of distribution and migratory movement of the Ural saiga population:

—

Zhanybek District—a traditional habitat area where large aggregations of animals are recorded;

—

Kaztal District—a zone of active saiga movement bordering Russia;

—

Bokey Orda District—one of the most significant habitats of the population, where regular migrations are observed.

The red boundary delineates the area where the main population of the Ural saiga is concentrated, which is the largest in Kazakhstan and in the world. This population accounts for more than 90% of the total species and regularly crosses transboundary territories, including adjacent regions of Russia (Astrakhan and Volgograd regions).

The rising population and expanded range of saigas in West Kazakhstan are intensifying their effect on the economy, leading to a more acute problem of pasture and hayfield degradation. To develop effective management strategies, it is essential to consider the trends in saiga population dynamics. From 1981 to 2023, the distribution of saigas has fluctuated significantly, reflecting the influence of various anthropogenic and external factors. Between 1981 and 1995, saiga populations remained relatively stable with only slight fluctuations. However, from 1995 to 1997, the population experienced a sharp decline due to factors such as poaching, habitat loss, and other human activities. In the years that followed, the population remained low, but steady growth was observed starting in 2015.

The West Kazakhstan region serves as a crucial distribution area for the Uralsk saiga population, which has experienced an eightfold increase over the past five years.

Figure 2 shows the dynamics of the three main saiga populations:

—

The Ural population (Ural)—inhabits western Kazakhstan, including the border areas with Russia;

—

Betpakdala population—widespread in central Kazakhstan;

—

Ustyurt population—located in the south-west of the country near Uzbekistan.

In 2018, the population was recorded at 217,000, but by 2023, it surpassed 1.9 million. This remarkable growth makes it the largest saiga population in the world.

The restoration of the Uralsk zone was made possible by improved security measures, a ban on hunting, the restoration of pasture ecosystems, and favorable climatic conditions. These factors have not only stabilized the population but also contributed to its record growth.

The West Kazakhstan region is home to the largest saiga population, making it crucial for the conservation of this species. However, the rapid growth of the saiga population presents new challenges, such as increased pressure on pastures, competition with livestock, and heightened risks of disease outbreaks. To address these issues, it is essential to develop scientifically sound population management strategies that balance sustainable development with the conservation of the region’s ecosystem resources.

The category ‘Saiga population numbers’ shows the total number of individuals of the species in Kazakhstan, combining all three populations. This data comes from official UNEP/CMS reports from 2021 and field observations from 2024.

Figure 3 illustrates a clear downward trend in pasture productivity across the Bokey Orda, Zhanibek and Kaztalov districts from 2014 to 2023. The most substantial decline is observed in the Kaztalov district, where productivity decreased from approximately 580 kg/ha in 2014 to about 200 kg/ha in 2023, representing a reduction of more than 66%. A similar negative dynamic is evident in Zhanibek, where productivity fell from around 680 kg/ha to 280 kg/ha (−59%). In the Bokey Orda district, yields decreased from roughly 750 kg/ha to below 420 kg/ha (−44%), indicating the most severe impact relative to the initial baseline.

The simultaneous and consistent decline in productivity across all three districts indicates a systemic degradation process rather than random fluctuations. The most probable driver is the rapid increase in the size of the Ural saiga population and the concentration of large herds along traditional migration routes. This leads to intensified grazing pressure, trampling of vegetation, and delayed recovery of plant biomass. These results highlight the urgent need for management interventions, including the implementation of pasture rotation, restoration programs on degraded lands, and the development of strategies to mitigate conflict between wildlife conservation and agricultural production.

Supporting this trend, the manuscript illustrates the spatial differentiation of pasture and hayfield productivity within the West Kazakhstan region.

Despite the moderate increase in average productivity at the regional scale, the three examined districts demonstrate abnormally high losses, sharply contrasting with neighboring territories. Therefore, the present study focuses specifically on these districts as critical degradation zones, which is consistent with the broader regional context.

Research was conducted to assess the agroecological impact of saiga on the sustainability of pasture ecosystems in the West Kazakhstan region, with a particular emphasis on biometric indicators during the 2024 agricultural season. The findings indicate that saiga exert a significant influence on the condition of pasture vegetation in the semi-desert zone, altering both quantitative and qualitative characteristics of forage resources.

Spring field surveys demonstrated that vegetation cover on regulated pastures reached 70%, whereas on areas with unrestricted saiga presence (control sites), it declined to 40%. Similar differences were observed in the height of the grass stand:

—

Under unrestricted grazing by saiga, the average height was 20.20 cm;

—

On regulated pastures, it reached 30.70 cm.

Since forage productivity is a key parameter for agricultural use, the green biomass yield was compared. Spring measurements showed that under rotational grazing, the productivity of green biomass reached 6.15 t/ha, whereas under uncontrolled saiga grazing, it was only 2.15 t/ha. The reduction amounted to 4.00 t/ha, which is equivalent to a loss of 41.8%.

Pasture types in the semi-desert zone differ in the composition of their phytocenoses; however, grazing pressure remains the dominant determinant of pasture condition. Regulation of population numbers should be aligned with the maximum carrying capacity of pastures based on valuable forage species such as Agropyron desertorum, Stipa capillata, Festuca valesiaca, and Leymus ramosus, and should not rely on the dominance of secondary species (Koeleria cristata, Kochia prostrata), which are less resilient to grazing pressure.

According to the research data from that year, the highest productivity, along with elevated feed and energy-protein values, was achieved in pastures where grazing of farm animals was regulated to prevent saiga invasion and to organize pasture turnover. In this case, the dry weight output reached 2.61 kg/ha, with the collection of feed units and digestible protein amounting to 1.45 and 0.15 c/ha, respectively, while providing digestible protein of 105 g per feed unit. Additionally, the yield of exchange energy from pasture feed was also higher, measuring 1.98 GJ/ha.

The excessive presence of saigas negatively impacts the overall productivity and nutritional value of pasture ecosystems. In the course of the conducted research in the semi-desert zone of the West Kazakhstan region, where up to 80% of the Ural saiga population is concentrated, a decrease in dry matter yield of 1.55 t/ha was recorded, along with reductions in feed units and digestible protein of 1.07 and 0.13 t/ha, respectively.

As a result of the saiga population’s effects, the availability of feed units containing digestible pasture protein decreased from 105 g to 58 g. Furthermore, the collection of metabolic energy fell from 1.98 gigajoules per hectare to 0.60 gigajoules per hectare, resulting in a decline of 1.38 gigajoules per hectare.

Biometric indicators of pasturelands, including grass stand height and biomass yield, were recorded during field surveys conducted as part of the agroecological assessment of the impact of saiga populations on pasture ecosystems in the West Kazakhstan region.

Measurements were carried out at experimental sites during the spring, summer, and autumn periods, with two main variants distinguished: pastures with regulated grazing of agricultural animals, where no saiga impact was observed, and pastures with free grazing of saigas (control).

As shown in

Table 2. Livestock population in farming households (heads).

District	2017	2018	2019	2020	2021	2022	2023
Cattle (Cattle Stock)
Bokey Orda	45,707	52,387	51,376	54,388	61,146	65,556	63,336
Zhanibek	23,245	26,322	30,267	32,204	32,826	35,111	37,258
Kaztalov	38,297	43,932	46,801	50,295	58,938	65,134	60,724
Sheep and Goats
Bokey Orda	68,638	56,249	51,760	50,200	56,896	55,743	47,820
Zhanibek	34,754	39,204	40,218	36,117	36,128	33,989	42,649
Kaztalov	122,449	128,659	133,764	127,547	127,425	135,312	108,816
Horses
Bokey Orda	16,357	18,308	19,262	20,228	24,761	25,210	24,645
Zhanibek	9,125	11,044	12,714	13,284	15,103	17,937	18,545
Kaztalov	13,551	15,581	17,037	18,982	20,852	22,508	18,675

Data source: Statistical collection “Key indicators of livestock development in West Kazakhstan Region, 2017–2023.” [26].

, the dynamics of livestock numbers in the Bokeyorda, Zhanibek, and Kaztal districts between 2017 and 2023 reveal a steady increase across all major categories—cattle, sheep and goats, and horses. Despite adverse climatic conditions, pasture degradation, and a declining forage base, farming households continued to expand livestock production, reflecting the persistence of a productivist agricultural model focused on output growth and food security rather than ecological balance.

The dynamics of livestock numbers in the Bokeyorda, Zhanibek, and Kaztal districts during 2017–2023 demonstrate a steady upward trend despite objective challenges such as pasture degradation, climate variability, and competition with the saiga antelope population. This growth can be interpreted as a manifestation of the productivist logic, in which the primary goal remains the expansion of production volumes and the maintenance of food security, even under ecological constraints.

In the Bokeyorda district, the cattle population increased from 45.7 thousand head in 2017 to 63.3 thousand in 2023 (+38%), while the number of horses rose from 16.3 thousand to 24.6 thousand (+50%). A similar pattern is observed in the Kaztal district, where the number of cattle grew from 38.3 thousand to 60.7 thousand (+58%). Even under conditions of shrinking pastures and a declining forage base, farms continued to expand their herds, seeking to compensate for reduced profitability through production growth.

From the standpoint presented by Lee, Tin, and Gunawan [27], this trend reflects the inherent tension between productivism and post-productivism—two competing paradigms in agricultural development. Productivism prioritizes output maximization, while post-productivism emphasizes balance between economic objectives and the preservation of natural ecosystems. The Chinese experience demonstrates that an abrupt shift away from a productivist model is unrealistic in countries with growing populations and high social expectations. Similarly, in the analyzed regions of Kazakhstan, despite degradation processes, farmers are compelled to maintain or even increase production to meet domestic demand and sustain rural employment.

The increase in livestock numbers amid declining pasture productivity can also be seen as a form of demographically driven productivism. As the rural population continues to grow, agriculture remains the main source of income and food self-sufficiency. Consequently, the expansion of livestock herds is not so much a strategic decision as a forced response to ongoing social and economic challenges. This finding aligns with the conclusions of Chinese researchers who note that, under conditions of limited resources, countries with a large rural population strive to maintain equilibrium between short-term productivity and long-term sustainability.

From an economic perspective, the observed tendency results in an increasing load on the region’s natural resource capital and a decline in the efficiency of pasture use per unit area.

Therefore, the growth of livestock populations, combined with ongoing degradation processes, generates a critical pressure on land resources in the region. The analysis of the current state of pasturelands and livestock numbers in the Bokeyorda, Zhanibek, and Kaztal districts of the West Kazakhstan Region reveals significant exceedances of the maximum permissible grazing load established by the Order of the Ministry of Agriculture of the Republic of Kazakhstan dated April 14, 2015, No. 3-3/332 [21].

Table 3. Pasture areas in Bokeyordinsky, Zhanibeksky and Kaztalov districts of West Kazakhstan region (ha).

№	District	Available Pasture Area, ha	Required Pasture Area, ha	Pasture Deficit, ha
1.	Bokey Orda	875,701	1,423,140	547,439
2.	Zhanibek	728,888	823,072	94,184
3.	Kaztalov	1,455,906	1,609,195	153,289

Data source: field observations and IRN AP23486846 [24].

shows that in all three areas, the permissible load on pastures has been exceeded. This situation could lead to land degradation and a reduction in productivity.

An analysis of the data revealed that in the Bokeyordinsky district, the actual load on pastures exceeds the permissible limits by 62.5%. In the Zhanibek and Kaztalov districts, the excess is 12.9% and 10.3%, respectively. This overloading results in reduced yields from hayfields and pastures, as well as degradation of the vegetation cover.

The situation is worsened by the large population of saiga antelopes in these regions. Recent data indicates that there are approximately 1.9 million saigas in the West Kazakhstan region, and their numbers are on the rise. Saiga antelopes compete with livestock for pasture resources, putting additional strain on the ecosystem. As a result, peasant farms in these areas are particularly impacted and are experiencing a shortage of fodder for their livestock.

To better illustrate the economic implications of changes in pasture productivity and livestock growth, the dynamics of profitability (loss ratio) in livestock production among farming households are presented in

Table 4. Profitability (loss ratio) of livestock production in farming households, %.

District	2017	2018	2019	2020	2021	2022	2023
Bokey Orda	46.7	49.2	47.2	51.8	46.3	46.0	45.7
Zhanibek	−10	−23.4	33.4	33.3	53.9	53.8	53.7
Kaztalov	20.0	41.1	65.6	119.5	23.8	23.7	23.6

Data source: Statistical collection “Agriculture, forestry and fisheries in the Republic of Kazakhstan, 2017–2023” [22].

.

As shown in Table 4 the profitability of livestock production in the studied districts varied considerably across the years. The greatest fluctuations were observed in the Kaztalov district, where profitability peaked at 119.5% in 2020, primarily due to increased meat prices and export activity. In the following years, however, profitability declined as feed costs rose. In the Zhanibek district, losses were recorded in 2017–2018, followed by a steady recovery in profitability after 2019. The Bokey Orda district maintained a relatively stable profitability level throughout the analyzed period, indicating a resilient production structure despite limited natural resources.

These data confirm that changes in pasture productivity and livestock numbers directly affect the economic performance of farming households, amplifying disparities in their financial sustainability.

Table 5. Calculation of the live weight index and pasture yield index (base year—2014, 2014 = 100).

	Bokeiordinsky Yield	Bokeiordinsky Live Weight	Zhanibeksky Yield	Zhanibeksky Live Weight	Kaztalovsky Yield	Kaztalovsky Live Weight
2014	100	100	100	100	100	100
2015	98.28	100	98.53	100	98.67	110
2016	96.55	102.7	97.06	102.78	100	112.5
2017	94.83	118.92	105.88	111.11	89.33	112.5
2018	94.83	116.22	73.53	111.11	89.33	110
2019	70.69	113.51	72.06	111.11	74.67	110
2020	65.52	113.51	60.29	108.33	72	110
2021	56.9	110.81	57.35	105.56	70.67	105
2022	51.72	108.11	42.65	102.78	56	102.5
2023	34.48	105.41	41.18	100	56	100

shows Pearson’s coefficients reflecting the relationship between the number of saigas (independent variable) and hay productivity (dependent variable) in three districts: Bokeiordinsky, Zhanibeksky and Kaztalovsky.

To quantitatively assess the impact of changes in pasture conditions on livestock production indicators, an economic–ecological model was developed based on the analysis of the relationship between pasture productivity and the performance of livestock farms. Sheep were selected as the indicator species because their productivity is highly sensitive to the quality of natural grazing lands and quickly reflects changes in vegetation structure and biomass availability. This approach enables an objective assessment of pasture degradation caused by increased migratory pressure from saigas and allows for a quantitative estimation of the economic losses experienced by farming enterprises. The model presented below illustrates the relationship between the dynamics of the saiga population, pasture yield, and the production performance of sheep farming.

Combined baseline data: pasture yield (kg/ha), average live weight of sheep (kg), and Ural saiga population size (heads).

To assess the relationship between pasture conditions and sheep production performance, a live weight index and a pasture yield index were calculated (base year—2014, where 2014 = 100). The results are presented in Table 5.

The results (Table 5) show that changes in animal weight and the condition of the forage base develop in opposite directions, which has important practical implications.

In Figure 4, the live-weight index remains generally stable over the ten-year period. In the Bokey Orda district, the increase in live weight from 37 to 39 kg corresponds to a rise in the index from 100 to 105.

Pasture yield in the Bokey Orda district decreases from 580 to 200 kg/ha, which is reflected in the decline in the index from 100 to 34.5. This indicates severe degradation processes within the forage base: over the ten-year period, pastures have lost almost two-thirds of their productivity, with the most dramatic decline occurring after 2018.

Figure 5 shows the dynamics of the pasture yield index and the live weight index of sheep in the Zhanybek district. A multidirectional trend is shown: with a relatively stable level of live weight, there is a sharp decrease in pasture productivity.

The presented chart clearly demonstrates a pronounced divergence in the dynamics of the pasture yield index and the live weight index of sheep in the Zhanybek district over the period 2014–2023.

Pasture productivity declined from an index value of 100 in 2014 to 41.18 in 2023, indicating a loss of nearly 60% of the natural forage base. The most substantial decrease occurred after 2017, when productivity fell from 105.88 to 73.53, and then continued to decline sharply, reaching its minimum level by 2023.

Since 2018, the pastures have entered a phase of accelerated degradation, indicating that the ecological carrying capacity has been exceeded and forage resources are being depleted. In contrast to the decline in pasture productivity, the average live weight of sheep remains stable. The weight index remains within the range of 100–110 points: increasing to 111 by 2017, then gradually decreasing and returning to the baseline value (100) by 2023. No significant losses in live weight are observed despite the deterioration of pasture conditions.

In Figure 6, the Kaztal District demonstrates a less pronounced but still substantial decline in pasture yield: the index decreases from 100 to 56, corresponding to a loss of approximately 44% of the forage base. Despite a relatively more stable trend compared to other districts, degradation becomes evident after 2018, reflecting broader regional dynamics.

In contrast, the live weight index remains stable, fluctuating within the range of 100–113 points. Peak values were recorded in 2016–2017 (index 112.5), followed by a gradual decline to the baseline level (100) by 2023.

Across all three districts, a pattern typical of pasture degradation is observed: the pasture-yield index declines sharply and systematically, while the live-weight index remains relatively stable. This divergence indicates that livestock productivity is increasingly maintained through economically costly compensatory feeding strategies. As a result, production costs rise, system resilience decreases, and financial risks in regional sheep farming intensify.

Increasing grazing pressure resulting from the rapid growth of the saiga population leads to a decline in the natural forage base and forces livestock producers to compensate for feed deficits through the purchase of external feed resources. As a result, farms increasingly rely on commercially supplied feed inputs (compound feed, haylage, silage, and grain) to maintain livestock live weight and production levels.

As noted in recent research, the expansion of saiga numbers intensifies competition for pasture resources between wild and domestic ungulates, thereby escalating conflict with the agricultural sector and generating direct economic losses for farmers. According to Serikbayeva et al. [28], high concentrations of saiga herds in the steppe regions of Kazakhstan have led to widespread complaints from farmers concerning losses caused by trampling of pastures, destruction of crops, and reduced opportunities for hay production for winter feeding. The authors emphasize that, under conditions of natural forage scarcity, farms are compelled to transition to compensatory feeding strategies, substantially increasing the financial burden on regional livestock production.

Serikbayeva and colleagues [28] further note that the conflict between wildlife conservation and livestock production intensifies precisely because farms shift from natural feed resources to expensive purchased feeds, transforming the issue from an ecological challenge into an economic one. The authors argue for the need to develop management mechanisms to regulate population numbers, such as establishing an optimal size of the Ural saiga population at approximately 75,000 individuals, in order to mitigate pressure on pasture ecosystems and support agricultural sustainability.

The live weight of sheep is maintained at a stable level despite the deterioration of pasture conditions.

A completely different trend is observed in the pasture-yield index. In the Bokeyorda district, yield declined to 34% of the 2014 level; in the Zhanibek district—to 41%; and in the Kaztalov district—to 56%. These figures reflect severe degradation of pasture resources, particularly since 2018–2020.

Pasture degradation has reached critical levels, amounting to a 44–66% decline over the past decade.

A comparison of the two indices reveals a critical observation. Although the natural forage base has declined nearly twofold, the live-weight level has decreased by only 10–15% relative to peak values and remains almost unchanged compared to the baseline year.

The stability of live weight is maintained not by natural pastures but through increased reliance on purchased feed.

This indicates that farms compensate for the loss of natural forage through intensive supplementary feeding, which leads to higher production costs, reduced profitability in sheep breeding, greater dependence on external feed supplies, and increased vulnerability of the sector to price and climatic shocks.

The increase in feed costs is becoming the key hidden consequence of pasture degradation, directly affecting the economic resilience of livestock farms. Thus, the index-based analysis confirms that the condition of rangelands is the primary limiting factor for the development of sheep farming in the region. Without systematic measures for pasture restoration, production costs will continue to rise, profitability will decline, and the sector will face a risk of economic collapse in the medium term.

Figure 7 illustrates the changes in the average live weight of slaughtered sheep in peasant and farm households over the period 2017–2023 in three districts—Bokeyorda, Zhanibek, and Kaztalov. The linear trend lines make it possible to identify the general direction of change and evaluate the stability of the dynamics.

Linear regression of the average live weight per head of sheep slaughtered in peasant and farm households (kg).

1.

Bokeyorda District—Moderate decline with a weak trend. The trend line for the Bokeyorda district demonstrates a positive slope (y = 0.2848x − 534.67); however, the actual dynamics indicate that growth in live weight was observed only during 2014–2017, after which a consistent decline occurred—from 44 kg in 2017 to 39 kg in 2023. The coefficient of determination R² = 0.1193 reflects a weak fit of the linear model to the empirical data, suggesting that variations are driven primarily by short-term fluctuations rather than a stable long-term trend.

Conclusion: the dynamics of live weight are unstable, with a clear tendency toward decline after 2017.

2.

Zhanibek District—Stagnation and minimal trend. The trend line for the Zhanibek district (y = 0.0545x − 72.2) is nearly horizontal, which is confirmed by the extremely low coefficient of determination (R² = 0.0091), indicating an almost negligible explanatory power of the model. The live weight of sheep remains within the range of 36–40 kg, without any significant upward or downward shifts.

Conclusion: the stability of live weight is maintained primarily through compensatory feeding, and the long-term trend is virtually absent.

3.

Kaztalovsky District—pronounced decline following the peak. The Kaztalovsky district demonstrates the highest live-weight values among the three districts, reaching 45 kg in 2016–2017. However, after 2018, a distinct downward trend emerges, with the average weight decreasing to 40 kg by 2023. The trend line has a negative slope (y = −0.2364x + 520), indicating a long-term decline in productivity, while the coefficient of determination (R² = 0.1321) remains low, though higher than in the other districts.

Conclusion: the sector in this district has entered a phase of systemic decline in live weight after 2018. Despite inter-district variation, the chart clearly demonstrates a common regional pattern.

During 2014–2017, an increase in live weight was observed in two districts (Bokeyorda and Kaztalovsky). However, after 2018, all districts demonstrate a stable downward trend in live weight indicators. The linear trend models show weak explanatory power (low R² values), indicating the influence of external factors and compensatory feeding practices.

Key conclusion: the dynamics of live weight do not directly reflect the condition of natural pastures—actual weight levels are maintained primarily through the use of purchased feed. Nevertheless, after 2018, even compensatory feeding can no longer offset the decline in natural forage productivity, resulting in a gradual reduction in live weight across all districts.

Figure 8 clearly demonstrates a pronounced and systematic decline in the productivity of natural pastures across all three districts over the period 2014–2023.

The linear trend confirms a long-term downward trajectory, with exceptionally high coefficients of determination (R²), indicating a stable and predictable pattern of degradation.

• Bokey Orda District—the most severe degradation. Pasture yield in the Bokey Orda district decreased from 580 kg/ha in 2014 to 200 kg/ha in 2023. The decline is reflected by the linear trend model:

y = −43.091x + 87,422 R² = 0.9075

(3)

—

A high R² (~0.91) indicates a consistent, almost perfectly linear degradation process.

—

The yield declines by approximately 43 kg/ha per year.

—

Over ten years, the total reduction amounts to 65.5%, demonstrating a critical loss of natural forage productivity.

The district is experiencing a critical overload of pasture resources and a pronounced loss of productivity, fully matching the increase in grazing pressure and the disruption of rotational pasture management.

2.

Zhanibek District—Rapid degradation trend. The Zhanibek district demonstrates a significant decline in pasture yield from 680 to 280 kg/ha over the study period. The linear trend is represented by the equation:

y = −51.818x + 105,104 R² = 0.8958

(4)

—

The average annual loss of yield is approximately 52 kg/ha, which is faster than in the Bokeyorda district.

—

The total reduction amounts to 58.8% over ten years.

—

The high R² value (~0.90) indicates a stable and systematic degradation process.

A notable feature is the short-term peak observed in 2017 (720 kg/ha), followed by a rapid decline, marking the transition of pasture ecosystems into a phase of accelerated deterioration.

3.

Kaztal District—relatively more resilient, but degradation remains substantial. Pasture yield in the Kaztal district declined from 750 to 420 kg/ha over the study period. The linear trend is expressed as:

y = −41.273x + 83,914 R² = 0.9429

(5)

—

The highest R² value (0.94) indicates exceptionally consistent and predictable degradation dynamics.

—

The total reduction amounts to 44%, which is less severe compared to the other districts.

—

The average annual loss is approximately 41 kg/ha.

Although the district remains comparatively more productive, the degradation trend is persistent and reflects an irreversible long-term deterioration of the natural forage base under increasing ecological pressure.

4.

General Regional Conclusion. All three districts demonstrate an almost perfectly linear trajectory of pasture degradation (R² = 0.89–0.94).

Pasture yield has declined by 44–66% over the ten-year period, with an annual reduction rate of 41–52 kg/ha per year. The graphical analysis clearly indicates a turning point in 2017–2018, when degradation processes accelerated.

Thus, the decline in pasture productivity is not random but represents a systemic phenomenon, driven by:

—

Excessive grazing pressure and overload of pasture capacity;

—

Disruption of rotational grazing practices;

—

Climatic and soil-vegetation shifts;

—

Increasing anthropogenic pressure and unmanaged wildlife migration impacts.

This confirms that pasture degradation has reached a critical ecological threshold and requires scientifically justified management interventions.

The conducted analysis demonstrates that the rapid growth of the saiga population in key districts of the West Kazakhstan region has led to measurable declines in pasture productivity and has intensified pressure on agricultural systems. The application of correlation and regression models confirmed a statistically significant relationship between the growth of the saiga population and the decline in hay yield: increasing animal numbers lead to a reduction in pasture productivity. These results highlight the need for evidence-based management strategies that integrate ecological monitoring with regional agricultural planning, ensuring that population dynamics of wild ungulates are adequately considered in decision-making by local and national authorities.

4. Discussion

The findings of this study demonstrate a pronounced and systemic decline in pasture productivity in the Bokeyorda, Zhanibek and Kaztalov districts over the period 2014–2023, which coincides with the rapid expansion of the Ural saiga population. The regression analysis confirmed a statistically significant strong negative relationship (p ≤ 0.001; |βs| = 0.86–0.93) between saiga abundance and pasture productivity, indicating that grazing pressure from large migratory herds is a key driver of ecological degradation in the region. Although environmental variability and livestock dynamics were not included in the first-stage model, the strong magnitude of the coefficients demonstrates that saiga population growth alone can explain a substantial proportion of yield decline.

The divergence between the pasture-yield index and the live-weight index of sheep reflects an adaptive response by farming systems: livestock productivity is being maintained through economically costly compensatory feeding rather than natural forage availability. This phenomenon is consistent with international observations concerning human–wildlife competition for grazing resources. For example, Nyhus [2] and Schwerdtner & Gruber [12] found that increasing pressure from wild herbivores forces farmers to rely on external feed inputs, triggering costly compensation mechanisms and reducing agricultural resilience. Likewise, Serikbayeva et al. [28] document growing conflicts in Kazakhstan as saiga herds concentrate in agricultural zones, leading to trampling of pastures, destruction of crops, and increased winter-feed shortages.

The findings of the present study align with the threshold-effect concept described by Gren et al. [5] and Singh & Milner-Gulland [4], who argue that pasture productivity may remain stable until grazing pressure reaches a critical ecological carrying capacity, beyond which degradation accelerates sharply. The abrupt decline in natural forage productivity after 2017–2018 in all three districts corresponds to the period of the steepest increase in saiga population growth, suggesting the crossing of such a threshold. This highlights the need for management strategies based on measurable ecological capacity rather than uniform nationwide policies.

The study also reveals that despite rapid degradation of pasture ecosystems, livestock numbers continued to grow between 2017 and 2023, reflecting what Lee, Tin & Gunawan [27] describe as a productivist agricultural paradigm, where production expansion persists even under resource constraints due to social and economic pressures. This combination of livestock intensification and uncontrolled wildlife grazing creates a cumulative effect on pasture degradation and increases the financial vulnerability of farming households.

Overall, this study reinforces the need to integrate ecological monitoring with agricultural planning and population regulation strategies for wild ungulates. The results support the implementation of adaptive management frameworks that set optimal saiga population levels, expand compensated management instruments, and develop pasture restoration and rotational grazing programs. Without timely intervention, continued decline of the natural forage base could lead to a long-term reduction in livestock productivity and undermine rural economic sustainability in the West Kazakhstan region.

5. Conclusions

This study revealed that the sudden increase in the saiga population in certain areas of the West Kazakhstan region has significantly harmed agricultural producers’ activities. By employing correlation and regression modeling methods alongside field research data, we were able to quantify the extent of the damage.

Based on the calculations, the following management recommendations can be proposed:

—

Introduce adaptive regulation of saiga numbers through annually revised management decisions based on remote-sensing data, population monitoring, and pasture condition indicators.

—

Strengthen rotational grazing systems and restore degraded rangelands using reseeding, soil-protection measures, and temporary exclusion zones to reduce cumulative grazing pressure.

—

Apply direct and indirect protective measures, including temporary fencing, visual and acoustic deterrents, and the establishment of buffer zones near vulnerable crop fields.

—

Develop compensation mechanisms for farmers who incur the greatest losses.

—

Develop integrated monitoring systems (biometric data, drones, GIS) in districts with critical overloading.

Based on the analysis of sources and field observations, it can be noted that anthropogenic pressure on ecosystems should also be taken into account. Multifactorial pressure on the environment has a significant impact on both pastures and the development of wild fauna. Ultimately, this will allow for the preservation of biodiversity and the harmonious development of the environment.